Abstract
The majority of solid tumours develop hypoxia because oxygen demand exceeds oxygen supply. In association with this is the poor prognosis and response to therapy. In oncology, it is well appreciated that merging anatomical and functional information has a significant impact on the extent of disease and target volume delineation. Commercial image-fusion software packages are becoming available but require comprehensive evaluation to ensure reliability of image-fusion and the underpinning registration algorithms, particularly for radiotherapy. The present work seeks to assess such accuracy for a number of available registration methods provided by the commercial package ProSoma™. The molecular imaging modality of Positron Emission Tomography (PET), in conjunction with radio-labelled molecules that undergo chemical changes inside tumours as a result of the presence or absence of oxygen, has became a promising technique for the non-invasive quantification of tumour hypoxia. Herein the relationship between tumour hypoxia and vasculature geometry is considered using a novel mathematical approach, likewise the spatiotemporal distribution of a hypoxia PET sensitive tracer is determined. Representation of the oxygen distribution in 2-D vascular architecture using a reaction diffusion model enables quantitative relationships to be obtained, specifically between tissue diffusivity, tissue metabolism, anatomical structure of blood vessels and oxygen gradients. Similarly, tissue activity curves (TAC) are a potential key in providing information on cellular perfusion and limited-diffusion. In this thesis a development to the work of Kelly and Brady (2006) is described and verified, with a particular interest in simulating TACs of the most promising hypoxia PET sensitive tracer, 64Cu-ATSM.